Mining on Manifolds: Metric Learning without Labels

arXiv (Cornell University)(2018)

引用 56|浏览153
暂无评分
摘要
In this work we present a novel unsupervised framework for hard training example mining. The only input to the method is a collection of images relevant to the target application and a meaningful initial representation, provided e.g. by pre-trained CNN. Positive examples are distant points on a single manifold, while negative examples are nearby points on different manifolds. Both types of examples are revealed by disagreements between Euclidean and manifold similarities. The discovered examples can be used in training with any discriminative loss. The method is applied to unsupervised fine-tuning of pre-trained networks for fine-grained classification and particular object retrieval. Our models are on par or are outperforming prior models that are fully or partially supervised.
更多
查看译文
关键词
hard training example mining,pre-trained CNN,Euclidean similarities,manifold similarities,unsupervised fine-tuning,fine-grained classification,metric learning,object retrieval
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要